On Jan 31, 2010, at 2:28 PM, Walter Lenk wrote:

> I believe that there is a bug in the FB5 implementation of MKI$


Walter,

Thanks for the report.

I've written C code to product the results requested but I'm wondering  
what constitutes corrective behavior.  Different behavior by FB4 is  
sometimes wrong.

FB Help says defstr byte should produce the same result as chr$ and it  
doesn't ( chr$ produces the result Walter expects ). This result  
favors the contention FB4 is correct but I'll continue to research.

Brian S